PPI - the role of CDC in decumulation

The role of Collective
Defined Contribution
in decumulation
Registered Company Number: 04145584. Charity Number: 1087856 (England & Wales)

PLEASE VISIT OUR WEBSITE:
www.pensionspolicyinstitute.org.uk
OR CONTACT:
Danielle Baker
Head of Membership & External Engagement
danielle@pensionspolicyinstitute.org.uk
FOR FURTHER
INFORMATION ON
SUPPORTING THE PPI
We have been at the forefront of
shaping evidence-based pensions
policy for over 20 years.
The Pensions Policy Institute (PPI), established in 2001,
is a not-for-profit educational research Institute. We are
devoted to improving retirement outcomes. We do this
by being part of the policy debate and driving industry
conversations through facts and evidence.
The retirement, pensions and later life landscapes
are undergoing fast-paced changes brought about
by legislation, technology, and the economy. Robust,
independent analysis has never been more important
to shape future policy decisions. Each research report
combines experience with INDEPENDENCE to deliver a
robust and informative output, ultimately improving the
retirement outcome for millions of savers.
Our INDEPENDENCE sets us apart – we do not lobby
for any particular policy, cause or political party. We
focus on the facts and evidence. Our work facilitates
informed decision making by showing the likely
outcomes of current policy and illuminating the trade-
offs implicit in any new policy initiative
About the Pensions Policy Institute
By supporting the PPI, you are aligning yourself with our vision to drive better informed policies and decisions that
improve later life outcomes and strengthening your commitment to better outcomes for all.
Prev Next

John Upton
Policy Analyst
Tim Pike
Head of Modelling
Published by the Pensions Policy Institute
© December 2023
ISBN 978-1-914468-15-5
www.pensionspolicyinstitute.org.uk
Funding has been given to help fund the research and does not necessarily imply agreement with, or support for the
analysis or findings from the project. The PPI does not make recommendations as to the appropriate direction of future
policy, Instead, our work provides INDEPENDENT evidence to allow policy development to be well informed.
This report is authored by:
The role of Collective Defined Contribution in decumulation
This research report is kindly sponsored by:
An INDEPENDENT Research Report by the
Prev Next

Page
Executive Summary 1
Introduction 3
Chapter One – A History of Collective Defined Contribution (CDC) in the UK 4
Chapter Two – A model for a CDC decumulation scheme 8
Chapter Three – The sensitivity to assets and members 16
Chapter Four – Further challenges to overcome 22
Appendix One 29
References 32
Acknowledgement and Contact Details 34
What can the UK learn about other countries’
approaches to accessing DC savings?
Prev Next

Executive Summary
CDC pension schemes become a reality in the UK when the Pensions Schemes Act 2021 became law, and regulations
governing CDC schemes were set out. The legislation, however, is flexible enough to support other applications for
CDC schemes.
The UK implementation of decumulation CDC, as currently envisaged, is unlike implementations of similar
arrangements across Europe. The UK approach to scheme design is constructed to offer greater fairness between
members, at the cost of less predictable future benefit levels.
The designs implemented in other countries, such as The Netherlands, have raised concerns of intergenerational
fairness and the subsidising of groups within the schemes. Maintaining the aim of greater transparency and fairness
in the UK will lead to challenges for UK decumulation CDC schemes as they will need to operate without smoothing
mechanisms, such as buffers, and with a more limited time horizon of future liabilities than whole of life CDC
implementations.
UK decumulation CDC schemes should be able to fulfil their
objectives while operating under current design constraints
A decumulation CDC scheme offers longevity risk pooling between members, unlike individual drawdown, and a
higher degree of investment in growth assets than an annuity. Decumulation CDC schemes can therefore offer a
higher, if slightly more volatile, income for life than an annuity without the risk of fund exhaustion associated with
drawdown.
Scheme objectives will be a critical part of communication with members as these outline the scheme’s approach to
risk and therefore the potential for benefit adjustments, including cuts, which will affect members most.
The potential scheme offers an income level which offers a high opening benefit level, which broadly increases in line
with the Consumer Prices Index (CPI) [Figure E.1], though the actual year on year increases vary [Figure E.2]. This
is achieved with a significant proportion of growth assets which offers a higher benefit level at the cost of reduced
benefit predictability.
This report explores the trade-offs in how a decumulation Collective Defined Contribution
(CDC) pension scheme may operate under the environment currently envisaged in the UK.
Figure E.1
A scheme may see increases or decreases in terms of real benefit
Deciles of the index of the real benefit adjusted for CPI of the base scheme
0
100
200
300
400
500
600
700
800
2028
2030
2032
2034
2036
2038
2040
2042
2044
2046
2048
2050
2052
2054
2056
2058
2060
2062
2064
2066
2068
2070
2072
2074
2076
2078
Benefit
index
(real)
Year
70%
60%
50%
40%
30%
20%
Example
1%
99%
Figure E.2
The modelled scheme provides future benefit increases within CPI ±2%
in two out of three annual valuations
Benefit increases payable by year for different percentiles of stochastically generated CDC scenarios
-30%
-20%
-10%
0%
10%
20%
30%
40%
50%
60%
2028
2030
2032
2034
2036
2038
2040
2042
2044
2046
2048
2050
2052
2054
2056
2058
2060
2062
2064
2066
2068
2070
2072
2074
2076
2078
Benefit
increase
payable
Year
90%
80%
70%
60%
50%
40%
30%
20%
Example
1%
99%
PPI: The role of Collective Defined Contribution in decumulation
1
Prev Next

The majority of risk and volatility issues in CDC decumulation
stem from investment performance
A CDC decumulation scheme will need to consider within its objectives the balance of a higher opening benefit level
and the degree of predictability of future benefit levels that can be provided. Higher initial benefit levels can be
achieved through growth-orientated investment strategies, though these are associated with greater degrees of
investment volatility and, ultimately, greater unpredictability in future benefit levels. Alternatively, using an investment
strategy which foregoes greater investment in growth assets, such as a cashflow driven strategy, it is possible to
generate a more predictable future benefit level, though starting benefit levels will be lower.
Cashflow-driven investment strategies are becoming more prevalent
in matching Defined Benefit (DB) liabilities and may contribute to the
investment strategy to ensure the product functions as intended.
The needs contained within the investment objectives to provide predictability in the future benefit payable to
members may result in investment strategies that are consistent with those observed in the more mature DB market.
Cashflow-driven investment strategies make greater use of hedging assets to reduce volatility, which, in turn, can
deploy dynamic discount rates that deliver greater funding stability. This approach necessitates a move away from
growth assets which would have a consequential impact in the level of benefit offered as the price of offering greater
predictability.
Without consistent underwriting there may be systemic unfairness in a
decumulation CDC scheme.
Within the fairness considerations of a scheme, it is imperative that the sharing of longevity risk be fair. Because
the longevity risk is a factor in the volatility of future benefit increases, it would need to be an equivalent risk for all
members. That is, their personal longevity risk when compared to the aggregate longevity risk within the scheme must
be consistent with other members. To achieve this would require an equivalent degree of underwriting for all new
members or else the volatility of future benefit levels (which would be applied uniformly) would not reflect the degree
of mortality risk (which would not be uniform).
Insufficient scale is more of a threat to the economic viability of
a decumulation CDC scheme than to its effective risk pooling
Risk sharing between members still operates effectively at a scale below which a scheme may not be economically
viable. The benefit level offered by very small schemes offer will inevitably be less predictable than that offered by
larger schemes, all other things being equal. However, the major driver to the scheme’s volatility remains investment
performance, regardless of scale, as the mortality volatility is lower than investment volatility and consequently has a
smaller impact on predictability.
For a scheme to be of sufficient scale to be economically viable it must be able to implement a charging structure
allowing it to cover running costs, as well as recover start-up costs including authorisation fees.
The minimum economically viable size will depend upon both upon the number of members and the total assets under
management. The necessary scale may be equivalent to that required for a master trust to have an adequate charging
base to cover its costs.
Decumulation CDC will not operate in a vacuum; interaction
with other decumulation options, both as competition and
compliment, will determine its success within the spectrum of
such products
Decumulation CDC will face competition from existing decumulation products, new products within the current
framework including blended solutions, alongside additional potential competition from multi-employer CDC schemes.
Income volatility in private pension income will be diluted by income
from other sources, such as the State Pension.
For any pensioner they will need to understand the degree of uncertainty they can tolerate from their retirement
income. Ultimately the desirability of retirement product mixes will depend upon the risk profile of the individual
member. The impact of income volatility from private pension decumulation options is diluted by the State Pension,
which provides a predictable income underpin with funding risk (in place of investment risk in a pay-as-you-go
scheme), inflation risk and longevity risk borne by the state.
Communication is the single biggest challenge to overcome
Experience from the Netherlands, where communication with CDC scheme members failed to align expectations with
the realities of the systems, resulted in reduced trust in their pension system and, ultimately, was a contributory factor
to the reforms which will come into force as the new Dutch pension agreement.
While the communication may be the single biggest challenge, this is still only an input to understanding the choices
in maintaining an income in retirement. A problem William Sharpe described as the “nastiest, hardest problem in
finance”.
Introducing decumulation CDC to the pension market will make a
complicated decision more complex.
It will be important to consider where decumulation CDC is positioned within the advised and guided markets and
how decumulation CDC should be positioned between the certainty of an annuity and the flexibility of an individual
drawdown product.
To ensure the best outcomes for individuals it may be appropriate to use a personalised mix of decumulation
products.
Since freedom and choice was implemented combinations of drawdown and annuity products have been proposed
which have become more sophisticated and tailored over time, with the aim of meeting consumers’ needs, including a
desire for flexibility.
The decision to use a decumulation CDC scheme is unlikely to completely answer the needs of a new member.
A personalised blend of products would enable an individual’s particular needs regarding income levels, need
for predictability and risk appetite to be accommodated. It is a complex balance to define the most appropriate
combination for an individual and advice may play a role in enabling this.
3

Pension pots ‘can be used to buy Lamborghinis’, says minister The Guardian, 20 March 2014
2
Prev Next

Introduction
Chapter One – A brief history of CDC in the UK
This chapter examines the place of CDC Pension schemes within the UK pension system
and how the current situation may be adapted to accommodate CDC decumulation
schemes in future
Chapter Two – A model for a CDC decumulation scheme
This chapter details the implementation of a model for a CDC decumulation scheme. It
covers the key features of the scheme and the critical assumptions that drive the model
outcomes.
We particularly consider the approach to setting key assumptions:
• Mortality
• Investment portfolio
The approach to liability management through future benefit increases allows the
scheme to manage the balance of assets against liabilities.
It summarises how the scheme performs under these conditions
Chapter Three – The sensitivity to assets and members
This chapter examines the impact of how particular sensitivities to investment portfolios
and membership impact the observed outcomes for the sample scheme.
We consider the level of benefits offered and the predictability of future benefit levels.
Chapter Four – Further challenges to overcome
This chapter examines other issues associated with the implementation of CDC
decumulation schemes in practice. Many issues have been raised within the 2023 DWP
consultation ‘Extending Opportunities for Collective Defined Contribution Pension
Schemes’.
Collective Defined Contribution (CDC) pension schemes become a reality in the UK when the Pensions Schemes
Act 2021 became law, and regulations governing CDC schemes were set out. The first scheme to be authorised is a
whole-of-life Defined Benefit-style (DB) scheme offered by Royal Mail [outlined in Box 1].
Box 1 - The Royal Mail CDC Scheme
Royal Mail’s Collective Pension Plan comprises a DB Lump Sum section, accruing at 3/80 of pensionable pay plus
increases, and a CDC section, accruing at 1/80th pensionable pay plus increases. The contribution rates are 13.6%
for the employer and 6% for the employee. Average increases are expected to be the Consumer Prices Index
(CPI) + 1% (but not guaranteed).
The legislation, however, is flexible enough to support other applications for CDC schemes. The PPI has previously
modelled and published work focussing on CDC during the accumulation phase,1 2
and in understanding what
lessons could be learnt from other countries who already have CDC.3
It is felt that little has been done to explore
the post-retirement space specifically to answer: How could a CDC-style decumulation-only solution work, what
would be the pros and cons and the likely member outcome?
Box 2 – Accumulation, decumulation and whole of life
Pension schemes and products operate across a life course. The period when a member is paying money
into a scheme in the form of contributions is the accumulation phase. This typically ends at retirement when
a member stops putting money in and starts taking it out to provide an income in retirement. This period
of withdrawing pension savings in the form of an income (or lump sum) is the decumulation phase. Getting
between accumulation and decumulation may require transferring pension savings between products and
providers. A scheme that operates over both accumulation and decumulation without the need to switch
products is referred to as “whole of life”.
Considering a scheme where members transfer a fund into a scheme to purchase an income for life, the PPI has
analysed the income that a scheme could provide to its members.
This research has set out to answer the following questions:
What could be the role and main design features of CDC in decumulation?
• What level and rate of indexation could be set for the target benefit level?
• What is the likelihood of this target being met?
• How might benefit cuts from underfunding impact projected benefits?
• How might any bonuses from overfunding get converted into benefits?
• What are the trade-offs between risk (such as investment policy) and outcomes (such as volatility)?
What are the constraints on effective operation?
• Assets under management (AUM);
• Number of members;
How could some of the other possible issues be handled? e.g.:
• Communication risk;
• Cross-generational cross subsidy; and
• What might an endgame look like.
1
Popat et al. PPI (2015)
2
Wilkinson, PPI (2018)
3
3
Prev Next

CHAPTER ONE:
A BRIEF HISTORY OF
CDC IN THE UK
This chapter examines the place of Collective Defined Contribution
(CDC) pension schemes within the UK pension system and how the
current situation may be adapted to accommodate CDC Contribution
decumulation schemes in future.
Prev Next

4
Pension Schemes Act 2021 ss 9 and 60
5
Pension Schemes Act 2015, pt 2
6
CWU (2018)
7
Pitt-Watson et al, RSA (2020)
8
Pensions Scheme Act 2021
9
TPR (2022)
10
Hansard (2020)
11
12
Hurman (2023)
13
14
Tversky Kahneman (1992)
For CDC pension schemes to operate as a decumulation-only product it is necessary to consider how the whole-of-life
CDC schemes have developed in the UK market.
CDC has been implemented in the UK providing an alternative
occupational pension scheme to the historical dichotomy
between the collectivism of Defined Benefit (DB) schemes and
the individualism of Defined Contribution (DC) schemes
The Pensions Scheme Act 20214
gained royal assent on 11 February
2021 passing the necessary legislation to allow the creating of CDC
schemes in the UK.
Legislation was first passed for Defined Ambition (DA) schemes in 2015,5
however the provisions never came into
force.
The act created the provisions for the Government to allow the creation of DA pension schemes. However, there was
no demand for the implementation of such a scheme and the necessary secondary legislation was never enacted.
In 2018, the Communication Workers Union (CWU) and Royal Mail Group (RMG) reached an agreement on the design
of a new CDC pension scheme. This was to replace the then existing DB and DC schemes open to employees of
RMG.6
To implement this proposed scheme necessitated legislation to be passed.
Rather than revisit the previous act, it was determined to be more expeditious to introduce new legislation to facilitate
the proposed CDC scheme. Within the legislation, clause 47 gives the Government powers to allow the creation of
multi-employer schemes, however the legislation was primarily designed to permit the creation of the RMG proposed
CDC scheme.7
Regulation of CDC schemes falls under the remit of The Pensions Regulator (TPR)
The Pensions Act 2021 sets out the authorisation regime for any new CDC scheme [Box 3]. The scheme is assessed
by TPR as to whether they are satisfied the scheme meets the authorisation criteria.8
TPR has a supervisory role in the
ongoing function of all CDC schemes.
Box 3 – CDC legislation and regulation in the UK
The Pension Schemes Act 2021 provides the legislative framework to establish and operate CDC schemes
(referred to as Collective Money Purchase (CMP) schemes in the Act) in the UK. The Act also provides for TPR to
produce a Code of Practice for the authorisation and supervision of CDC schemes.
TPR consulted on its proposed Code of Practice between January and March 2022. It came into force 1 August 2022.
The Code of Practice covers the statutory objectives of TPR and considers the authorisation of schemes under
the six authorisation criteria set out in legislation:
• Fitness and propriety
• Systems and processes
• Member communications
• Continuity strategy
• Financial sustainability
• Sound scheme design.9
Rather than include reference to fairness in the act the Parliamentary Under-Secretary of State for Work and
Pensions, Guy Opperman, explained fairness was best considered in regulation.10
A decumulation CDC product regulated by TPR would be in competition in the market with products regulated by the
Financial Conduct Authority (FCA), creating a regulatory distinction between the product and its competitors.
UK CDC schemes are regulated to operate on a no buffers, best estimate basis to mitigate risks of unfairness
between scheme members. This approach is based upon the demographics of a whole-of-life scheme with a broad
membership base and a long liability profile across which to mitigate the impact of variable investment returns and
mortality experience.
The alternative international implementations, to include buffers and reserving [Box 4], improves the predictability of
future benefit payments. The Dutch do not make adjustment to future benefits when the funding ratio fits within a
funding gate. Further, their use of recovery plans reduces the likelihood of making benefit cuts.11
In the decumulation-
only space, German variable-life annuities are adjusted according to the scheme funding position if the funding ratio
moves outside of 100% to 125%, while Dutch variable pensions have the ability to smooth a member’s benefit level
over a period of up to ten years which is directly applied to the member’s capital value (“smoothing with your own
pot”) to limit intergenerational risk sharing.12
These approaches could necessitate the adoption of funding regulation in
the UK (regulated by TPR) as this could bring them into line with DB schemes.
Box 4 – Features of international CDC implementation
Buffers: Capital buffers are assets set aside in a CDC scheme to mitigate against the need to vary future benefit
levels. Where the scheme is underfunded, assets from the buffer are used to maintain benefit levels and, where
the scheme is over funded, the buffer may be replenished rather than distributing all of the surplus.
Funding Gate: A scheme cannot distribute any surplus or cut benefits in the case of a deficit when the scheme
is only under or over funded by a small margin, e.g. ±5%. This avoids making so many changes to benefit levels,
however when an adjustment is necessary it may be larger.
Reserves: Money set aside to cover future adverse experience, see ‘buffers’.
Predictability of the benefit paid by UK CDC pension schemes is predicated on the long-term nature of membership.
Adjusting future benefit increases over this longer period has a significant impact on the value of future liabilities. In
the decumulation-only space, CDC schemes would not be able to spread lower than expected returns across a broad
member base and so, all other things being equal, members would be more likely to experience reduced indexation
and potentially nominal cuts to pensions in payment, as risk is shared amongst a smaller group.13
Research has shown that people experience twice as much pain from a loss as pleasure from a gain of equal size,
which may mean that pension funds would seek to avoid delivering outcomes below people’s expectations.14
In
terms of scheme design, where expectations of gains and losses are set on a best estimate basis resulting in an
approximately equal chance of gain or loss, it may be necessary to seek to control the scale of these gains or losses.
This may be the only recourse to mitigate this pain which could result in distrust and disappointment in a CDC
arrangement, alongside careful management of expectations through communications.
5
Prev Next

The Royal Mail has authorisation for the first CDC scheme in the UK.
The Pensions Scheme Act 2021 was designed to enable the implementation of the Royal Mail pension scheme and,
hereafter, it may be considered that this scheme may operate as a template for future UK pension schemes.15
Clearly,
implementing a decumulation-only product to operate in the competitive retirement market will require going beyond
the structure of the Royal Mail scheme, however it is anticipated that it will remain built upon the same principles.
The Royal Mail Group scheme is designed to provide the best affordable pension outcome for Royal Mail employees.16
It is designed to operate for a particular membership and at a particular cost.
The investment strategy backing the scheme effectively transfers investment risk from older members to younger
members via the annual pension increases.17
This risk transfer is what facilitates the pension smoothing as the future value of younger members’ benefits is
far more sensitive to changes in the rate of benefit increases. This sensitivity is effectively used to mop up the
volatility of investment returns without having such a significant impact upon benefit increases for older members.
This allows younger members who may benefit from accepting the risks in the scheme to take it from older
members who are less likely to be in a position to benefit from risk taking. In a decumulation-only situation, this
risk transfer will not be available as younger members, many years from retirement, are not present in the scheme.
The Royal Mail has a workforce of 150,000 people18
which offers a large supply of active members to their pension
scheme.
This offers the scheme an immediate scale of membership, albeit assets will start at zero and grow over time as
contributions are made. Most importantly to the scheme design, it generates a broad membership base across which
to share the investment risk between younger members and those approaching retirement. It future, the membership
will age to include pensioner members who will be able to share risk with new, younger members.
Other potential occupational schemes.
It is estimated that for an employer to be best placed to operate a CDC scheme that is financially viable, they may
need a workforce of at least 5,000 employees.19
This is significantly below the size of the Royal Mail Group.
A prospective membership of 5,000 employees translates to over 100 active members at every individual year of
age. Below this scale it may be necessary to consider a multi-employer arrangement to achieve sufficient scale for
adequate risk pooling.
The advantage of a single employer scheme is that the scheme does not need to operate in a commercially
competitive environment. That is not to say it does not need to deliver value to its members, this is to be monitored by
TPR, but that the offering can be tailored to the members and the active membership will remain stable, subject to the
ongoing stability of the employer.
CDC is being considered as a potential pension solution to
improve outcomes for individuals for the same contributions
Modelling has shown potential whole-of-life CDC retirement outcomes may be at least 30% higher when compared
to an individual arrangement.20
This earlier modelling tends to reflect a scheme design more akin to international
implementations with a buffer operating through a funding gate approach (by virtue of predating the legislation that
prescribed CDC scheme design in the UK). Regardless, the additional value was still generated largely through an
investment strategy more heavily based upon growth assets whose volatility can be mitigated through investment risk
sharing between younger and older members.21
This advantage over an individual arrangement is still available within
UK scheme designs, however the outcomes would be expected to be more volatile with the restriction upon buffers.
It could be that this increased volatility may lead to an investment strategy in practice that is less growth seeking to
offset investment volatility, which would in turn reduce the scale of benefit from being within a CDC arrangement (over
an individual arrangement).
CDC schemes are being considered for whole-life multi-employer
applications.
This may necessitate some modifications to the single employer scheme design as currently legislated, however some
features will remain the same. Notable considerations that may also be meaningful in the decumulation space include:
• Sectionalisation of schemes;
• The need for competition and assessment of benefit levels and levels of risk;
• TPR oversight of communication and marketing;
• The use of Technical Actuarial Standards to provide illustrations and in marketing;
• Maintaining an approach of ‘no buffers’ when undertaking scheme valuations to assess the funding position; and
• Bounds on adjustments to future indexation.
It is indicated that multi-employer schemes, if implemented, could be based upon the same legislative and regulatory
framework as Master Trusts.22
Multi-employer schemes would allow access to a CDC scheme for
employers who do not have a suitable employee profile to support
their own scheme
15
16
CWU (2018)
17
Donnelly, IFOA (2022)
18
Royal Mail Group (2023)
19
Eagle et al, WTW (2020)
20
Popat et al (2017), Aon (2020), GAD (2009), Pitt-Watson et al, RSA (2012), Eagle et al, WTW (2020)
21
Taylor, Ward (2023)
22
DWP (2023)
6
Prev Next

In supporting transparency and fairness, UK CDC scheme design may
produce more volatile outcomes than international implementations,
where volatility is managed through buffers using a funding gate
mechanism.
UK whole-of-life CDC scheme designs share risk across a broad membership base, using longer time
horizons to produce more predictable benefit levels.
In decumulation only CDC schemes the time horizon is significantly shorter, reducing the efficacy of
benefit increase mechanisms to smooth benefit levels.
Decumulation CDC schemes would need to operate in a commercially
competitive environment.
Schemes would be competing for new members and, for this effective competition, it will be necessary
to communicate how benefit levels and levels of risk may compare not just to other decumulation CDC
schemes, but also to alternative decumulation products.
CDC schemes tend to produce better-modelled outcomes than
individual arrangements through greater use of return-seeking
investment strategies and pooling of mortality risk.
Where there is less opportunity to share investment risk across membership in decumulation-only CDC
schemes it may be necessary to use a more conservative investment strategy than their whole-of-life
counterparts. This may result in a reduction in the advantage that a decumulation-only scheme may be
able to achieve while maintaining a risk exposure suited to its membership.
Conclusions
7
Prev Next

CHAPTER TWO:
A MODEL FOR A
COLLECTIVE DEFINED
CONTRIBUTION (CDC)
DECUMULATION SCHEME
This chapter details the implementation of a model for a Collective
Defined Contribution (CDC) decumulation scheme. It covers the key
features of the scheme and the critical assumptions that drive the
model outcomes.
We particularly consider the approach to setting key assumptions:
• Mortality
• Investment portfolio
The approach to liability management through future benefit increases
allows the scheme to manage the balance of assets against liabilities.
It summarises how the scheme performs under these conditions.
Prev Next

More detail on the design of the model and the assumptions used are contained in the Modelling Appendix. Results
for sensitivities and other scenarios are included in Chapter Three.
The approach to member mortality
To model the scheme, we assume the scheme receives a constant supply of new members. They are aged 68 at
the beginning of their membership and the pots they bring to the scheme are based upon the distribution of pots
currently used to purchase annuities. The number of new members is set intentionally high so that the impact of
mortality risk pooling functions as intended in this base case.
Members age and die according to a random stochastic process.
Each year, the model picks randomly and independently for each member whether they die in that year. The
probability for each member is based on two things: how old they are, and when they were born. This is used to
build a personal mortality table for every individual, which is taken from population statistics and allows for mortality
improvement over time.23
For example, 90 years olds have a higher chance of dying than 70-year-olds, and someone
who is 80 in 2030 has a higher chance of dying than someone who is 80 in 2050, because it is projected that life
expectancy will increase over time.
The probability of death for each member is independent of any other member. Evaluating these probabilities
randomly, running the model multiple times leads to a set of distinct stochastic results. Modelling one cohort of 1,000
members who all join in 2028, we see that the number of surviving members in each year after that follows an S
shaped curve [Figure 2.1], where the distribution is fairly tight: the difference in population between the 1st percentile
and 99th percentile of stochastic runs of the mortality of a 1,000-person cohort is never more than 75 lives in the
entire lifetime of the cohort.
Figure 2.1
Even with random variation, the mortality of 1000 68 year olds in
2028 is relatively predictable
Spread of members left alive in each year in 1000 stochastic runs of a model CDC scheme with one cohort of 68
year olds joining in 2028
Mean
1%
99%
-
100
200
300
400
500
600
700
800
900
1,000
2028
2033
2038
2043
2048
2053
2058
2063
2068
2073
2078
Number
of
members
alive
at
start
of
year
Year
9th Decile
8th Decile
7th Decile
6th Decile
5th Decile
4th Decile
3rd Decile
2nd Decile
The scheme is structured to include new members at the beginning of
each year and the membership takes approximately 30 years for the
membership to stabilise.
Within the base scheme membership every cohort that joins will be at retirement age. At first, everyone in the
scheme will be relatively young and it will only be once the first cohort has mostly died out that the age distribution
will stabilise, with every age group being represented. This takes approximately 30 years when all new members join
at age 68. The proportion of people at each age stays roughly the same, assuming roughly equal numbers of new
members each year.
It takes around 30 years for the membership to become stable, and the scheme to be matured
We see that, if we model a scheme where a cohort of 1,000 people join each year, in the first 15-20 years, the total
size of the membership grows fairly linearly. This linear growth can be attributed to low mortality levels across
everybody in the scheme, given that nobody in the scheme is yet old enough to have a particularly high probability
of dying. After this initial linear growth, the earliest cohorts start to reach the ages where their probability of dying in
a given year becomes high, and the growth of the total membership starts to plateau [Figure 2.2]. Eventually, after
about 30 years, the membership stabilises completely: for each year where 1,000 new 68-year-olds join, there will be
approximately 1,000 deaths in the preceding cohorts.
Figure 2.2
It would take approximately 30 years for a scheme’s membership to
reach a stable age profile
Age of members in the scheme by number of members by 5 year tenure brackets
-
5,000
10,000
15,000
20,000
25,000
2028
2033
2038
2043
2048
2053
2058
2063
2068
2073
2078
Number
of
members
alive
at
start
of
year
Year
40+ years
35-39 years
30-34 years
25-29 years
20-24 years
15-19 years
10-14 years
5-9 years
1-4 years
New joiners
The average future life expectancy of the membership decreases from 19.9 years when the first new members
arrive, to a minimum of 13.2 years when the membership stabilises 32 years later [Figure 2.3]
Number of members alive at the start of the year
original single cohort of 999 lives
Number of members alive at the start of the year by
membership length 1000 new members each year
23
ONS (2022)
9
Prev Next

The term over which benefits can be smoothed though risk pooling can be represented by the average life
expectancy across all members in the scheme. It represents the number of years the average member will live for.
It decreases from scheme inception until the scheme becomes mature [Figure 2.3]. Initially, the scheme is purely
populated by 68-year-olds who have the largest future life expectancy of any members (being the youngest). As
the membership ages their future life expectancy reduces. The minimum future life expectancy is after 32 years
and represents a reduction in life expectancy of the membership by one third. It is only after this, with a stable
membership, that the average future life expectancy starts to increase in line with future mortality improvers.
Figure 2.324
The outstanding average life expectancy decreases as the
scheme matures
Weighted life expectancy of members Constant rate of new members aged 68, ONS 2020-based mortality
for females
-
5
10
15
20
25
2025
2030
2035
2040
2045
2050
2055
2060
2065
2070
2075
Weighted
life
expectancy
across
scheme
membership
Year
The life expectancy observed in a decumulation-only scheme is far below the future aggregate life expectancy than
could be expected in any open whole-of-life scheme where the membership would be dominated by working-age
individuals in a mature scheme.
We effectively assume that the membership, on average, is
underwritten perfectly fairly and consistently.
The model uses the same life tables25
for three distinct purposes in the base realisation (though all may be set
independently):
1. Setting the benefit level for new members on a best estimate basis. The mortality is used to calculate an
opening benefit level for a new member (i.e., the pricing basis);
2. Calculating the best estimate value of future liabilities for any existing member on a best estimate basis. The
mortality is used to calculate the value of the liability for any member (i.e., the valuation basis);
3. Calculating the probability of death of a member in the scheme is projected in any year. The mortality is used
to designate particular members as dying or surviving in any year of projection (i.e., the experience basis).
By using the same mortality rates for each purpose, the model implicitly assumes that lives have been perfectly
underwritten. Where the pricing basis is systemically out, this would imply ineffective underwriting and may result in
particular new members either subsidising or being subsidised by other members of the scheme. Where the valuation
basis is systemically out, the value of future liabilities will be misvalued, leading to a bias in future benefit adjustments.
Where the experience basis is systemically out, the scheme will find itself consistently under or over funded.
In practice, a scheme would not have a perfect estimate of future mortality rates, and the mortality estimate for a
new member would be updated on a best estimate basis as that member continued in the scheme. Where there is a
bias to the estimation of mortality, the scheme will tend to penalise those with either a short lifespan or long lifespan
depending upon the direction of the bias.
The approach to setting an investment portfolio has to be
tailored to meet the objectives of the scheme
The portfolio of the backing assets is split 40:40:15:5 between
equities, gilts, bonds and cash, with volatility reduction to reflect
meeting an investment goal whereby assets are selected to more
closely correlate with the liability profile.
This portfolio allows the scheme to deliver adequately consistent
benefit increases while including growth assets to be able to achieve
objectives of a CDC scheme.
The benefit adjustment that a member receives at the end of each year is directly linked to investment performance
in that year. Although other factors affect their benefit to a lesser extent, such as the number of members who die
and the number of new members who join with whom risk can be shared, investment performance has the capacity
to cause large fluctuations in the benefit adjustments applied each year. Within a CDC scheme the benefit is not
predetermined at outset: it is not bound in the same way that a Defined Benefit (DB) scheme would be to provide a
particular benefit. However, broadly speaking, a CDC scheme should still aim to provide a consistent, reliable benefit
increase in order to give members financial stability and peace of mind. This creates a trade-off between providing
some level of stability, while also utilising the capacity to accept risk to seek higher return and therefore higher overall
benefit.
The scheme’s ability to spread the investment risk across members is reduced when compared to whole-of-life CDC
schemes. This is due to the shorter average future membership period (13.2 years) when compared to a whole-of-life
scheme.
Aggregate life expectancy
24
PPI analysis of data from ONS (2022)
25
ONS (2022)
10
Prev Next

More volatile, return-seeking asset mixes would lead to greater uncertainty in future benefit levels (albeit with a higher
overall return). The objective of the model scheme was picked such that the benefit increase deviating by more than
2% from the Consumer Prices Index (CPI) was restricted to a one in three chance.
Ultimately, it was determined that a portfolio mix of 40:40:15:5 between equities, gilts, bonds and cash would result
in this scheme’s objective being met under the modelled central conditions and membership demographic. The
objective is, in a sense, arbitrary and reflects a scheme’s willingness to accept risk in the pursuit of higher benefits. It
will determine the scheme’s investment strategy and is a balance that the trustees of any new scheme would have to
grapple with.
This portfolio is more return seeking than portfolios typically designed
to back annuities or DB schemes and so offers a higher lifetime
income than other guaranteed-for-life products or schemes.
The assets used to back an annuity portfolio can vary, but one rule of thumb suggests that they should be made up
of 90% bonds.26
The asset mix used in the decumulation CDC scheme offers a higher return and therefore a higher opening benefit
level. With the proposed assets profile of 40% equities, a member who bought in with a £100,000 pot could expect
an opening benefit of £6,077, anticipated to rise with inflation. Modelling the same CDC scheme, but with the 10%
equity investment more comparable to an annuity, this same member would see an opening benefit of £4,890, which
is 19.6% lower and does not include a margin for risk premium or the cost of capital associated with an annuity.
For a £100,000 pot the scheme would offer an initial benefit level of
£6,077 per year anticipated to rise in line with increases in CPI.
The benefit level is set such that the value of the future benefit and charges is valued to be the same as the
pension pot transferred into the scheme.
When a member joins the scheme, their initial benefit level is set so that, if they live as long as they are expected
to, if inflation grows as it is projected to, and if investments perform as they are expected to, then in theory they
will receive in benefits exactly the same value as they paid into the scheme when they joined (less charges payable
towards scheme expenses). Of course, in practice these estimates and projections will not be completely accurate,
but this is the approach used to set the initial benefit for a member.
This initial benefit is lower than currently available level annuity rates, however the CDC benefit still offers a higher
initial income level than currently available from an index-linked annuity.27
If the member dies before they were projected to, the value of the benefit they were paid will be less than the
value of the pension pot they transferred into the scheme. The difference is used to support other members who
may live longer. Longevity risk is shared between members, those who die early financially support those who live
the longest.
Longevity risk is the risk that somebody only budgets for living for a certain amount of time, and then lives longer
than anticipated, thereby running out of money. In the case of somebody who retires with a Defined Contribution
(DC) pension pot, they might estimate that they will die in 15 years, and then each year draw down 1/15th of their pot.
If they die 10 years after retirement, they will have some of their pot left over, and could have had a higher income
each year. However, if they die 20 years after retirement, they will have spent their money too quickly, and will have
to stretch what little they have left once they realise they will live longer than initially budgeted for. This is particularly
problematic as, on average, retirees’ total household spending per person remains relatively constant in real terms
through retirement, increasing slightly at ages up to around age 80 and remaining flat or falling thereafter.28
Any fall
in income may result in pensioners no longer being able to sustain this rate of expenditure and experiencing a fall in
living standards.
A CDC scheme reduces this risk for a member by sharing this risk between all members. The scheme will estimate
when each member is likely to die, and calculate their benefit accordingly so that it is affordable for the scheme.
However, this estimate is just an estimate, and any given member could still die sooner or later than anticipated.
The difference is that, by joining together in a CDC scheme, they protect those that live longest: the ones who die
sooner than anticipated may not “make their money back”: what they paid into the scheme may be more than they
get back out of it, because they did not live long enough to collect enough payments. This money that is contributed
by members who die before they can claim it all back (as there is no further benefit paid on or after the member’s
death) is used to cover those members that live longer than anticipated: their benefit will have been calculated on the
assumption that they would die sooner than they actually did, and therefore the benefit they claim in their lifetime will
be worth more than the initial pot they bought in with.
This does not eliminate the risk entirely: if too many members live longer than anticipated, then there will still not be
as much money to go around as expected. However, given that mortality is fairly predictable on a large enough scale,
joining together in a CDC scheme vastly reduces the longevity risk for any individual member, giving them the peace
of mind that no matter how long they live, they will have an income that is fairly steady.
10% of 2028’s new members within the modelled scheme are projected to live less than nine years, while at the
other end of the spectrum 10% of 2028’s new members are projected to live for at least 29 years.
At the extreme ends of this distribution, even when considering that every individual in a cohort has the same
mortality risk, we still see that the last 10% of the cohort to die will claim their benefit for roughly three times as long
as the first 10% to die. This demonstrates well the deal that an individual member is making with all other members: in
this example, a member who lives nine years and a member who lives 29 years both had the same chances of dying
in each year, and could not have anticipated that they would live especially short or long lives. In joining the scheme,
they agreed to trade the possibility of not gaining all their money back for the almost guaranteed income for life. A
member who lives only nine years essentially contributes a large amount of their pot to providing this stable benefit
for the members who live 29 years.
The approach to balancing assets and liabilities
When the scheme’s assets end up being higher or lower than expected at the end of the year (based on a best
estimate projection), the scheme makes an adjustment to the benefit. This adjustment determines the benefit that
a member receives in the following year, but also sets the expectation for the rate that the benefit will be uprated
every year in the future, until another year of unexpected performance causes the scheme to update this adjustment
again. This adjustment factor is the measure of how volatile the member’s benefit will be. If the adjustment factor
were always zero, then a member would see their benefit uprated exactly alongside inflation, which would be as
stable a benefit as possible and consistent with the benefit level provided by an annuity. The model bounds the future
adjustment to the rate of benefit increases at ±2% from CPI increases, and, when this bound is applied, makes a one-
off benefit adjustment to balance assets and liabilities.
A scheme must realise any shortfall in its asset position every year.
Using a median investment return to calculate the value of assets means that there is a 50:50 chance that future
benefit increases may rise or fall as a result of investment performance
In a similar way to how estimates of future mortality rates are used to calculate a member’s benefit level, so too
are projected investment returns. Rates of investment return are more volatile than mortality experience, and so
in practice a scheme will have to be prepared for investment performances that are significantly better or worse
than anticipated. This can be mitigated with defensive investment approaches such as a cashflow-driven investment
strategy, however this will tend to forego growth assets and, as a result, higher overall benefit levels would have to be
foregone. It will be necessary within a scheme to strike a balance when using defensive investment strategies which
prioritises predictability over total benefit level.
The model uses the median expected portfolio rate of return as a best estimate rate used to value future liabilities.
What this means is that, within the model, there is a 50% chance that investment performance will be better than the
anticipated performance used in the valuation of a member’s benefit, and a 50% chance that it will be worse.
26
Blanchett Finke, (2018)
27
Comparison is made with current best annuity rates taken from Hargreaves Lansdowne (2023)
28
Crawford et al. (2022)
11
Prev Next

In this model, and indeed in a UK decumulation-only CDC scheme as it would be legislated currently, any investment
return that is above or below what was expected for that year, will be directly reflected in the benefit level of the
next year through adjustments to the benefit increase. The scheme cannot increase the benefit by a conservatively
small amount so that some of the extra investment return from years with a higher-than-expected return can be held
back to smooth the benefit in case of poor returns in subsequent years; the scheme must factor in all assets when
increasing the benefit and continue with the assumption that all future years will have the same investment return as
predicted. The use of such buffers has been explicitly prohibited in current permissible UK CDC scheme designs.
In the long run, this should lead to an appropriate benefit level for the scheme, since there is a 50% chance of the
investments under or over performing. However, this can lead to large fluctuations in the benefit in particularly volatile
periods of investment return, where benefit adjustments are down one year and rebounding the next.
DB schemes have faced the perennial challenge of asset volatility when publishing their funding position.
It has led to increasingly defensive asset allocation strategies [Figure 2.4]. A number of DB schemes have active and
deferred members below pension age which means they may have a longer liability profile than a decumulation-only
CDC scheme. This should result in them being less affected by investment volatility as they are able to mitigate asset
and investment volatility over a longer period. Yet still, those schemes that had leveraged liability driven investment
(LDI) strategies, found themselves to be vulnerable to market activity in September and October 2022.
Figure 2.429
DB schemes asset allocations have become increasingly defensive to
reduce asset and investment volatility
Percentage of assets by asset class for funded DB schemes
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2006 2008 2010 2012 2014 2016 2018 2020 2022
%age
allocation
to
asset
class
Year
Equities Bonds Other Investments
The value of scheme assets is influenced by the volatility of
investment returns on the fund.
The scheme asset position at the end of each year is based upon three primary cashflows: pots transferred in by
new members, which should be balanced by a corresponding movement in liabilities; benefit pay out, which should
be predictable over the course of one year; and investment returns (net of charges), the most unpredictable of the
cashflows. Together these determine the asset position of the scheme in each simulation [Figure 2.5].
Figure 2.5
The asset balance of a CDC Scheme is most predominantly affected
by investment performance
Future cash flows of a model CDC scheme by year, in a single example of a stochastically generated economic
scenario
£(500,000,000)
£-
£500,000,000
£1,000,000,000
£1,500,000,000
£2,000,000,000
£2,500,000,000
£3,000,000,000
2025
2035
2045
2055
2065
2075
Annual
cashflow
Year
New pots
Investment return
Benefit outgo
Assets
Liabilities
Scheme cashflows (nominal)
29
PPF (2022)
12
Prev Next

The value of future liabilities depends upon the rate of future benefit
increases.
When assets turn out to be higher or lower than expected, then the benefit for that year will be uprated by CPI plus
or minus an adjustment factor. This adjustment factor is calculated so that, in every subsequent year, under best
estimate assumptions, each member’s benefit can continue to be uprated by CPI plus or minus this same adjustment
factor. The adjustment factor is identified which, with a precision of 0.1%, most closely matches the liabilities to the
assets of the scheme [Figure 2.6].
Where the combined effect of the increase in CPI and the adjustment would result in a nominal benefit cut, there
is no nominal benefit cut applied in the first year. These may otherwise manifest when inflation is negative or the
adjustment when applied to best estimate future CPI increases results in a negative projected benefit increase (e.g., a
CPI increase of 1% combined with an adjustment of -1.5%). The scheme instead places a floor on the benefit change, in
effect a double lock of price inflation with benefit adjustment and zero.
Figure 2.6
The benefit of each member is valued each year in order that the total
scheme liabilities match the total scheme assets
Example valuation for a scheme with one member
Benefit payable
during year
Closing assets Potential
future benefit
adjustment
Benefit payable
for next year
Liability Benefit increase
for future years
£414.57 £5,766.91 CPI + 1.3% £422.30 £5,679.11
CPI + 1.4% £422.72 £5,730.78 CPI + 1.4%
CPI + 1.5% £423.13 £5,783.08
Effectively there is a 50:50 chance whether the increase available will be better or worse than the current best
estimate being used.
Because the model predicts the investment return of any given year to be the median investment performance in that
year, there is an even chance of this adjustment factor being positive or negative.
To prevent distortions the long-term future benefit increases are set
within the range of CPI ±2%.
There is a consensus that the inclusion of a limit to long-term future benefit increases is appropriate when applied to
whole-of-life schemes.30
These bounds were proposed in the Government consultation and respondents recognised
that this would result in new members joining a scheme with a target rate of future benefit increases not too
significantly deviating from CPI.31
Under current monetary policy, the Bank of England is responsible for managing
inflation with a target rate of 2%.32
Under these conditions the lower bound would represent a benefit amount which is
level (CPI at 2% with a benefit adjustment of -2%).
When the value of future liabilities cannot be made to match the value of assets by using a future benefit increase
rate in this range, an additional one-off benefit adjustment is applied.
If the adjustment factor would need to be greater than ±2% to balance scheme assets and future liabilities, then the
adjustment factor for long-term benefit increases is set at this bound. On top of this, to ensure that the value of future
liabilities matches the scheme assets, a one-off benefit adjustment factor, which scales the next year’s benefit level
and is subsequently carried forward into future benefits, is applied [Figure 2.7]. This one-off adjustment factor is
calculated as the value of scheme assets divided by the value of future liabilities (calculated with the long-term benefit
adjustment at the bound).
Figure 2.7
The adjustment is capped at 2%, with any additional shortfall in
liabilities made up by a one of extra benefit adjustment
Example valuation for a scheme with one member
Benefit payable
during year
Closing assets Potential
future benefit
adjustment
Benefit payable
for next year
Liability Benefit increase
for future years
£340.70 £6,192.24 CPI + 1.9% £347.44 £5,144.98
CPI + 2.0% £347.78 £5,414.58
CPI + 2.0% £414.57 £6,192.24 CPI + 2.0%
When a bound has been struck there is only a 50% chance of leaving it in the next year – whether the best
estimate return is bettered or worse.
If the adjustment applied has reached a bound (±2%) the probability of returning to a more central rate of benefit
increases in the subsequent year is approximately 50%. This reflects the probability of the actual investment returns
being higher or lower than the best estimate median return. So, for every year in which there is an additional benefit
adjustment is made there is a 50% chance there will also be a need for an adjustment to be made again in the next
year. This can distort the number of boundary case adjustments observed where an additional benefit adjustment will
need to be made despite the actual return not varying from the best estimate by a large amount.
New members are only priced to the future benefit indexation of CPI ±2%
This places bounds on the pricing impact of the scheme’s past performance. If a scheme were to try and price for a
very high rate of future benefit increases this would lead to a very low opening benefit amount, and vice versa.
To avoid this unintended impact on a scheme’s operation, including its ability to attract new members, it is proposed
to realise gains or losses that cannot be accommodated by the bounded future benefit adjustment by making a
one-off alteration to benefit levels. This ensures that the range of benefit increase rates that would be offered by
competing decumulation CDC schemes would be broadly consistent, and schemes would not find themselves offering
excessively high or low opening benefit amounts resulting from the adjustment to future benefit increases. Though the
difference in opening benefit offered with an adjustment of ±2% could still be around ±20% from a central benefit level.
New members can only be priced accordingly in order to prevent schemes from offering higher, but more volatile
starting benefits to gain a competitive advantage.33
30
DWP (2023)
31
Grant (2023)
32
BOE (2023)
33
Grant (2023)
13
Prev Next

The operation of this baseline scheme under these conditions
The modelled scheme generates future benefit increases in the range
CPI ±2% in 68% of future years.
The rate at which the benefit adjustment is outside of these bounds, therefore, is a good indicator of how predictable
the future benefit stream is for a member. A scheme that must frequently adjust the benefit by more than 2% will
provide a retirement income to members that is less stable, including potentially severe benefit reductions that could
have implications for members’ ability to pay for regular life expenses. To assess a scheme’s stability, the ratio of
benefit adjustments made over the years that are outside this 2% range can be measured. For the scheme design
proposed in this paper, targeted adjustments are outside of this range to be incurred approximately one third of the
time.
When we consider the impact of one-off adjustments beyond ±2% we observe material benefit adjustments in years
where investment returns have been either particularly high or low. In the most extreme deciles of outcomes in any
year we see adjustments around 5% or greater from CPI, with even greater variation in the tails of the distribution
[Figure 2.8].
Figure 2.8
The modelled scheme provides future benefit increases within CPI
±2% in two out of three annual valuations
Benefit increases payable by year for different percentiles of stochastically generated CDC scenarios
-30%
-20%
-10%
0%
10%
20%
30%
40%
50%
60%
2028
2030
2032
2034
2036
2038
2040
2042
2044
2046
2048
2050
2052
2054
2056
2058
2060
2062
2064
2066
2068
2070
2072
2074
2076
2078
Benefit
increase
payable
Year
90%
80%
70%
60%
50%
40%
30%
20%
Example
1%
99%
If a scheme can provide some perception of stability to the member by not having too many large fluctuations in
the benefit, another key factor in how the benefit will be perceived is the likelihood of making a nominal benefit cut.
The avoidance of cuts, especially cuts to the nominal value of the benefit, would be a distinct but related goal to that
of keeping a stable benefit level. For the base scheme identified in this report, the nominal benefit rises steadily on
average, with it being very rare for the nominal benefit to ever dip below the opening benefit, [Figure 2.9]. The real
benefit, after adjusting the nominal benefit for CPI, is more equally likely to increase or decrease year on year, as this
reflects the best estimate for future benefit increases [Figure 2.10].
Figure 2.9
A scheme will see steady increases in nominal benefitin a large
majority of scenarios
Deciles of the index of the nominal benefit of the base scheme
0
200
400
600
800
1,000
1,200
2028
2030
2032
2034
2036
2038
2040
2042
2044
2046
2048
2050
2052
2054
2056
2058
2060
2062
2064
2066
2068
2070
2072
2074
2076
2078
Benefit
Index
(nominal)
Year
70%
60%
50%
40%
30%
20%
Example
1%
99%
Figure 2.10
A scheme may see increases or decreases in terms of real benefit
Deciles of the index of the real benefit adjusted for CPI of the base scheme
0
100
200
300
400
500
600
700
800
2028
2030
2032
2034
2036
2038
2040
2042
2044
2046
2048
2050
2052
2054
2056
2058
2060
2062
2064
2066
2068
2070
2072
2074
2076
2078
Benefit
index
(real)
Year
70%
60%
50%
40%
30%
20%
Example
1%
99%
The scheme can achieve its objectives with a carefully managed portfolio designed to reduce volatility and a
sufficiently large pool of members to mitigate the longevity risk.
14
Prev Next

Decumulation CDC Schemes can choose more return seeking assets
than other lifetime income products.
Because there is no promise to deliver a fixed income, a CDC scheme can afford to accept more risk
and deliver a higher overall benefit to members as a result.
Taking on too much risk will result in an unstable member benefit.
A member’s benefit increase each year is highly dependent on recent investment performance.
Although a more return-seeking, volatile investment portfolio for the scheme could lead to a higher
overall benefit across retirement, it would also cause more variation in the member’s benefit year on
year.
A scheme where everyone joins at retirement age will take a long time
to mature and have a stable age distribution.
Until it can reach a point where the first cohort of joiners has largely died out, a CDC scheme will have a
membership that is younger than in the scheme’s long term stable state. Given current life expectancy
estimates, this would take approximately 30 years.
Conclusions
15
Prev Next

CHAPTER THREE:
THE SENSITIVITY TO
ASSETS AND MEMBERS
This chapter examines the impact of how particular sensitivities to
investment portfolios and membership impact the observed outcomes for
the sample scheme.
We consider the level of benefits offered and the predictability of future
benefit levels.
Prev Next

A more return-seeking investment strategy results in higher
opening benefit amounts, which can vary by 38% across a 25%
to 75% equity investment strategy
If a scheme starts with a more return-seeking asset mix, this will be factored into the pricing basis which employs best
estimate assumptions. Since the scheme will be expecting higher returns ahead of time, the opening benefit is always
higher for a more return-seeking portfolio. The opening benefit can range from £5,500 per £100,000 of pension pot
for a scheme with 25% of its assets invested in equities, to £7,600 for a scheme with 75% of its assets invested in
equities [Figure 3.1]. Lower growth portfolio selections can be made bring the asset mix closer to the backing assets
for annuities where predictability is paramount.
Figure 3.1
A more return-seeking portfolio leads to a higher opening benefit
Opening benefit of schemes with different proportions of their assets invested in equities
£0
£1,000
£2,000
£3,000
£4,000
£5,000
£6,000
£7,000
£8,000
£9,000
75%
70%
65%
60%
55%
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Opening
benefit
per
£100,000
pot
Proportion of portfolio as equities
Opening benefit
In a competitive market, a higher opening benefit level could make a more attractive proposition to members. This
could lead to a competitive desire to focus on higher benefit levels at the cost of predictability.
With greater investment in return-seeking assets the volatility of portfolio returns is increased, leading to a less
predictable benefit level to a member. This may be more preferable, generating a higher overall income depending
on the member’s needs and other financial provision, or may be less preferable where a member is unable to absorb
income volatility. Using the asset mix selected for the base scheme presented in this paper – specifically, 40% equities,
40% gilts, 15% bonds and 5% cash – a balance is struck between the increased return from the freedom of a Collective
Defined Contribution (CDC) to invest more ambitiously, and the stability that a member may be able to accept in
their benefit level. This portfolio split offers a spread of investment returns under different stochastically generated
economic scenarios, which, as well as offering a reasonable proposal for a CDC scheme that aims to provide high
member satisfaction, gives a range of experimental conditions to be able to test the effectiveness of the CDC under a
wide range of investment performance scenarios.
A more volatile investment strategy results in less predictable
future benefit increases
By varying the volatility of the portfolio returns directly, rather than through adjusting the investment strategy, it is
possible to isolate the impact of investment volatility.
A one-third increase in the volatility of portfolio returns increases the
rate of one-off adjustments from 33.1% to 39.6%.
The result of using a more growth-seeking asset portfolio in a CDC scheme is that the investment return in any
individual year will be less predictable, as growth assets are associated with higher volatility. The benefit increase each
year is heavily dependent on the investment performance of the year prior to valuation and the consequential impact
upon the asset position of the scheme.
Isolating the impact upon the unpredictability of future benefit increases as a result of the investment risk is achieved
through manipulating the distribution of the rates of investment return. Increasing or decreasing return volatility
directly impacts the distribution of benefit increases. The impact is dampened to a degree by spreading over the
future period of benefit payments, however it remains the most significant driver to the predictability of future benefit
levels [Figure 3.2].
Figure 3.2
High volatility, even with the same overall return, leads to higher
volatility of member benefit
Effect on member benefit stability of artificially altering volatility by a factor of 1/3, while keeping mean returns the
same
Membership change % of years where adjustment was outside +-2% bounds
High volatility 39.6%
Base case 33.1%
Low volatility 29.0%
In reality, the volatility of portfolio returns is linked to the volatility in the returns of the underlying assets. Using a
higher proportion of growth-seeking assets (equities) leads to a higher volatility of the total portfolio return and a
greater proportion of annual valuations, where an adjustment of ±2% is inadequate to balance assets and liabilities
[Figure 3.3]. Approximately half of these adjustments represent a bonus to members, while half represent additional
cuts to benefits.
17
Prev Next

Figure 3.3
More return-seeking portfolios lead to higher levels of instability for
member benefit
The percentage of years in which an extra adjustment outside the 2% bound needs to be made by the proportion
of high-volatility equity investments
Below -2% threshold
Above +2% threshold
0%
5%
10%
15%
20%
25%
30%
75%
70%
65%
60%
55%
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Proportion
of
years
exceeding
threshold
Proportion of portfolio as equities
Considering the impact of the portfolio in greater detail [Figure 3.4] shows that the distribution of benefit adjustments
inherits the same skew as the distribution of portfolio returns. That is, a longer positive tail. Where the spread of
benefits is greatest, this is associated with the highest opening benefit levels offered [Figure 3.1]. This is the trade-off
of predictability versus benefit level that a CDC scheme would need to address when setting their scheme objectives.
Figure 3.4
Reducing the risk profile of the portfolio reduces the volatility in
member benefit adjustment
Spread of benefit adjustment deciles when using different amounts of assets as a proportion of the total
investment portfolio
-30%
-20%
-10%
0%
10%
20%
30%
40%
75% 70% 65% 60% 55% 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
Benefit
Adjustment
relative
to
CPI
% of portfolio assets in equities
9th decile
8th decile
7th decile
6th decile
5th decile
4th decile
3rd decile
2nd decile
1%
99%
While all members may benefit from a higher income, many may not
be able to accommodate the unpredictability that may be associated
with them.
The State Pension will provide a predictable underpinning income
It is unknown to what degree a member would rely on the income from a CDC scheme, however there is a growing
dependency on Defined Contribution (DC) pension savings with a corresponding drop in dependency on Defined
Benefit (DB) schemes. They could buy into multiple retirement products, possibly even multiple CDC schemes, and,
perhaps most crucially, they will receive some or even most of their retirement income from the State Pension. Those
with below median retirement income on average receive half of their income from the State Pension.34
Where a member requires greater predictability of income and is unable to sustain potential drops in income,
choosing the portfolio that delivers the highest return may not be appropriate. This model shows the sensitivity of
member benefit levels to investment performance in individual years. By choosing the most return-seeking portfolio,
members would be exposed to more uncertainty about their retirement income, which may not be worth the
potentially higher overall benefit. If a member is relying on this income to cover regular living expenses, the volatile
benefit that comes from volatile investments could have serious implications for their living standards.
Individual drawdown offers greater flexibility in managing investment
return volatility.
In a DC drawdown arrangement, an individual is essentially free to choose how much income they draw down and
when. This means that, as long as the member has several years’ worth of income remaining in their fund, they can
essentially manage their own regular income, taking what they need when it personally suits them, and independently
of the recent performance of their investments. This does come with greater risk, as investment returns may not
improve and make up for an experienced shortfall, leaving a member further exposed as a result of having withdrawn
more assets.
Decumulation CDC does not allow this flexibility because it risks intergeneration unfairness.
This freedom to draw a flexible income level does not exist in UK CDC design proposals. A member receives a regular
benefit of an unspecified amount, which is strongly linked to investment performance. Without buffers or reserves in
a CDC scheme, any volatility must be crystallised annually into the future benefit level. To not do so would place an
unequal burden on members which varies according to the length of their future life. This would result in unfairness
between differing cohorts and members within a scheme, and hence these features will not exist in the current
imagining of UK CDC.
Investment strategy and performance will be critical for any
decumulation-only CDC scheme achieving its objectives.
The modelling has shown that a member’s benefit is most heavily linked to investment performance. However, this
is not the only factor in determining member outcomes; the number of people coming in and out of the scheme
and impacting the demographics of the membership also affects the benefit paid. However, we see that any large
fluctuation in investment performance is immediately followed by a large benefit adjustment. The evidence suggests,
therefore, that investment performance will be critical to the final member outcome of a CDC scheme, both in terms of
overall benefit and the stability of that benefit. To understand exactly how crucial this is, it is also important to examine
the effects of any other variables.
34

Pike (2018)
18
Prev Next

Variation in the rate of new members joining the scheme is
secondary to investment risk in determining the predictability of
future benefit levels
If the model is run with extremely small numbers of members, this results in only a modest increase in instability of the
benefit. We see that even in an impractical example of one new joiner every five years, the benefit adjustment only
breaches the 2% boundary 36.7% of the time, compared to the 33.1% of the time that this would occur in the case of
100 members every year [Figure 3.5]. However, the scale of the more extreme swings is increased and associated
with poor mortality pooling. Given that this most extreme example of instability from small membership is still not
comparable to the potential increase in instability from choosing a more volatile asset portfolio, it can be concluded
that the effects of small membership are secondary to the effects of investment strategy.
Figure 3.5
Altering size of the membership of the scheme has little effect on the
stability of the benefit
Effects on benefit stability of varying membership size and rate of new members
Membership % of years where adjustment was
below -2% bounds
% of years where adjustment was
above 2% bounds
1000 members every year 17.9% 15.2%
1 member every year 18.4% 15.7%
100 members every five years 18.5% 14.7%
10 members every five years 18.8% 14.9%
1 member every five years 21.2% 15.2%
Longevity risk pooling operates effectively with any practical
membership size.
Given that extreme edge cases in terms of membership size do not affect the stability of the scheme, the minimum
size constraints imposed on a CDC scheme will instead be imposed by the costs of running the scheme that define
economic viability. The costs of setting up and starting to operate a new pension scheme are significant. For context,
the four largest master trusts have spent around £1bn between them in their first 10 years of operation (over the
period 2010 to 2019).35
There is also a fee for authorisation of £77,00036
for whole-of-life CDC schemes which is
calculated to cover the costs of the Pensions Regulator (TPR). Assuming a decumulation CDC scheme were to be
subject to the same charge alongside other initial costs, this would need to be accounted for when undertaking the
business case to implement a new scheme.
The main issue that arises from not achieving adequate scale is the ability of the scheme to be finically viable,
rather than being able to pool mortality and investment risk effectively.
Assuming annual charges of 0.75% of assets under management (AUM), a scheme with assets of £15,000,000,
which operates with functional mortality risk sharing, would generate an annual charge of £112,500. This scale would
not be sufficient for a master trust to operate, given that these costs would have to cover, among other things: the
costs paid to asset managers or an in-house team to manage the scheme’s funds; the costs incurred to implement
the investment strategy; the costs of advice taken on investments; the remuneration and overheads associated
with paying trustees and staff; the cost of professional services commissioned by the trustees, such as actuarial,
audit and legal advice; the cost of scheme administration, activities associated with receiving and reconciling
payments, compliance and communications to members; and any costs incurred by the employer, including planning,
communication, marketing and any dedicated support provided by them to the scheme.37
Since this would almost
certainly not be sufficient for a master trust, it would likely not be economically viable to run a CDC scheme on this
basis either.
For small schemes, economies of scale may require a larger pot per member to make them financially viable
The minimum average active pot size needed for a master trust pot to be economically viable was £4,100 in 2019.38
Schemes may implement minimum values for pots to ensure that they are economically viable by reducing the
proportion of their overheads which are fixed (generally administration costs) when a charging structure may be
based upon AUM.
An unstable number of new entrants does not significantly
disrupt the effectiveness of the scheme
A scheme in the commercial sector is likely to face fluctuating rates of new joiners. This may be in response to
reported fund performance and competitors spending on marketing. To model these effects a number of scenarios
have been modelled, reflecting increases and decreases in the number of new members. It was found that linearly
ramping up, linearly ramping down, and introducing a period of especially high or low membership in the middle of the
scheme’s life had little effect on its predictability of future benefit increases [Figure 3.6].
Figure 3.6
Altering the rate at which new members come into the scheme has
little effect on the stability of the benefit
Effect on benefit stability of changing rates of new members
Membership change % of years where adjustment was
below -2% bounds
% of years where adjustment was
above 2% bounds
Consistent rate of new members 17.9% 15.2%
Ramping down 18.4% 15.1%
Ramping up 16.8% 15.9%
Temporary drop in new members 17.9% 15.2%
Temporary increase in new members 17.9% 15.2%
These sensitivities alter the average duration of liabilities as the average member age is varied. It therefore follows that
the impact of new members joining at a wider variety of ages would also be secondary to the impact of investment
return volatility.
Where a scheme is economically viable it will have passed the threshold for membership above which there is
effective risk pooling to manage mortality risk.
35

Adams, PPI (2020)
36
The Occupational Pension Schemes (Collective Money Purchase Schemes) Regulations 2022
37
TPR, 2022
38
Pike, PPI (2021)
19
Prev Next

Investigation into more sophisticated investment strategies
could quantify improvements in benefit predictability
The modelling presented includes simplifications and limitations. This is reflected in the investment options considered.
This has consequences for the pattern of investment returns, particularly in the light of scheme objectives, that an
investment portfolio might generate.
Alternative portfolio options could include infrastructure and illiquid
funds as well as asset classes designed to hedge against investment
volatility.
The portfolio modelled is restricted to certain asset classes. Asset classes which are currently used in limited
quantities in pension products, but may be well suited to the needs of a CDC scheme, have not been included, and the
impact of these could be investigated in further work. This includes classes such as: infrastructure; property; private
markets; and commodities. Further, the use of hedging assets, which may be used to reduce the volatility of portfolio
returns by hedging against equity volatility, or to better align the asset returns to match the movements in the future
value of liabilities, could be considered as an effective investment strategy.
To account for some of this impact, the volatility of portfolio returns has been modified downwards in the results
presented.
The impact of asset- and liability-matching strategies could reduce
the volatility of the valuation of scheme liabilities.
Cashflow-driven investment strategies can make use of dynamic discount rates to value future liabilities. This links
movements in asset yields to the value of future liabilities, reducing the volatility of the funding position of a scheme.
This would reduce the volatility of benefit adjustments in a decumulation CDC scheme, generating a more predictable
future benefit level.
20
Prev Next

Investment volatility is the primary factor in member benefit stability.
We see that reducing volatility, without altering overall investment return, decreases members’ benefit
instability. This further highlights the trade-off that CDC investment managers will have to make
between return and stability, considering the needs of all their members.
The constraints on a scheme originating from membership rates are
secondary and driven by overhead costs rather than features of the
scheme itself.
We see that even in impractical illustrative examples, unusual membership conditions do not affect
scheme stability to the same degree as investment volatility. Other operational limits on the minimum
number of members are identified which trump any limit imposed by the scheme design itself.
Conclusions
21
Prev Next

CHAPTER FOUR:
FURTHER CHALLENGES
TO OVERCOME
Prev Next
This chapter examines other issues associated with the implementation
of Collective Defined Contribution (CDC) decumulation schemes in
practice. Many issues have been raised within the 2023 DWP consultation
‘Extending Opportunities for Collective Defined Contribution Pension
Schemes’.

Several additional challenges exist to the potential implementation of decumulation-only CDC schemes. These
are generally highlighted within the 2023 Department for Work and Pensions (DWP) consultation ‘Extending
Opportunities for Collective Defined Contribution Pension Schemes’39
and this chapter reflects upon a number of them
in a wider context.
Other product features require further consideration beyond the
scope of the modelling presented
The statutory right to transfer out is unlikely to be available to CDC
decumulation members.
The option to transfer out of a pension in payment in the UK is currently limited. In the March 2015 Budget, the
Government announced plans to permit a secondary annuity market40
Although these plans were cancelled in
October 2016 after consultation and engagement with the pension industry.41
Potentially a more valid comparison
could be made to the transfer out of a Defined Benefit (DB) pension scheme where the statutory right to transfer
benefits out is predicated on having one year or more years before the scheme’s normal pension age.42
That is to say,
those members currently in receipt of their benefit will not have a statutory right to transfer their benefit.
The Government stated a need to ensure protections apply to members in decumulation-only schemes, which was
deemed to be a significant stumbling block to the proposals of a secondary annuity market. The understanding of the
approach to annuities when considering decumulation-only CDC43
would indicate that the statutory right to transfer
out is unlikely to be available.
That is not to say that the scheme may allow transfers out or commutations at the discretion of the scheme trustees,
but there is the potential for buyer’s remorse owing to the nature of the product.
Without consistent underwriting there may be systemic unfairness in
a decumulation CDC scheme.
Within the fairness considerations of a scheme, it is imperative that the sharing of longevity risk be fair. Because the
longevity risk is a factor in the volatility of future benefit increases, it would therefore need to be an equivalent risk for
all members. That is, their personal longevity risk when compared to the aggregate longevity risk within the scheme
must be consistent with other members. To achieve this would require an equivalent degree of underwriting for all
new members, or else the volatility of future benefit levels (which would be applied uniformly) would not reflect the
degree of mortality risk (which would not be uniform).
Without fair underwriting there is the chance of adverse selection risk to the scheme. Members would be predisposed
to join the scheme where they hold the view that their mortality risk compares favourably to the pricing of benefits.
They would have an unfair advantage within the longevity risk-sharing pool, expecting a subsidy from those with
shorter life expectancies. Where a scheme is aware of the uneven distribution of this risk, this would reflect inequality
in their underwriting and an unfairness within the scheme.
Within the annuity market, mortality risk is borne by the insurer who may set their own risk preferences. Within a CDC
scheme, mortality risk belongs to the members and must be managed on their behalf by the trustees.
There are practical considerations that would need to be overcome, such as the cost and availability of appropriate
underwriting.
Single-life, index-linked payouts may not be the only benefit options
if decumulation CDC schemes are to provide true competition to
annuities.
If the decumulation-only CDC scheme is to be a competitor to annuity products it is natural to consider the options
that are available on the annuity market and consider whether these could be applicable to a CDC arrangement.
Options available on the annuity market include:
Guarantee periods: These ensure a minimum value of benefit payout to an annuitant in the case of death shortly after
taking out the policy. Within a CDC arrangement this would reflect a limit to the longevity risk sharing. As schemes
must be priced on a best estimate basis, variation between members on the degree of longevity risk sharing would
not be within the spirit of a CDC scheme. It could potentially instead be managed through a term assurance product
sold alongside the CDC scheme. The premium for the term assurance would need to be taken from the pot value of
the member as they transfer in, with the residual amount used to purchase CDC benefits.
Joint-life arrangements: For an annuity, this maintains the benefit payment until the death of the second member
(rather than some or all of the benefits terminating on the first death). This could be managed within a CDC
arrangement, as joint-life tables can be used for best estimate future mortality to enable risk-sharing calculations.
Analysis would need to be undertaken to ensure that such members would not present potential unfairness to other
members. Should it be necessary, it could be managed through sectionalisation of the scheme, subject to scale.
Impaired, enhanced and underwritten lives: Underwriting is used to assess the particular mortality of individuals and
price the benefit level accordingly. This is consistent with the philosophy behind setting initial benefit levels for new
members of a CDC pension scheme.
Alternative benefit escalation rates: Annuities may be sold with level, fixed, or indexed benefit increases. Within a
CDC scheme where the value of future liabilities is determined through adjusting the target rate of benefit increases,
it is imperative that all members are subject to the same treatment when setting future benefit increases to ensure
fairness. The investment objectives needed to meet varying benefit increases would similarly need to vary. It may be
possible to produce a sectionalised scheme to accommodate different indexation targets, though it may be necessary
to segregate members to ensure fairness when adjusting future benefit increases.
Implementing such options could come at an increased cost and increase complexity of the decisions a member
would need to take.
Smoothing benefit changes over multiple years.
Within current regulation for whole-of-life schemes is the option to apply a multi-annual reduction to benefits,
rather than realising a significant adjustment in one go. This could be used to smooth the impact of poor investment
performance over multiple adjustments, reducing the income shock members would otherwise need to absorb in one
year.
39
DWP (2023)
40
HMT (2015)
41
HMT (2016)
42
TPR (2023a)
43
DWP (2023)
23
Prev Next

A sophisticated investment strategy will need to match the scheme’s
objectives while delivering value for money to members.
In the event of a scheme opening, there is no doubt that the portfolio blend and investment strategy would be more
sophisticated than used in this illustrative modelling. The implemented investment strategy would be tied to, amongst
other things:
• Setting a competitive opening benefit amount
» Assuming that the rate of increase of benefit is predetermined by the scheme rules, adjustment to the
investment portfolio would be reflected in the opening benefit available to new members. The more growth-
seeking the investment strategy, the higher the opening benefit.
• Managing the predictability of benefit increases
» The investment strategy may be set to ensure that the predictability of future benefit increases. This would
require a cashflow-driven investment strategy to match the assets more closely to the liabilities.
» This is reflected in the value of future liabilities where the liabilities are discounted according to the best
estimate asset returns. Matching the growth in asset cashflows to the growth in liability cashflows reduces the
uncertainty in this valuation.
• Delivering value for money
» Decumulation products which are not defaults are outside of the current charge cap regime, however there is
an onus placed upon trustees by The Pensions Regulator (TPR) to assess value for money. Trustee boards are
expected to consider investment governance when assessing value for money for members.44
Cashflow-driven investment strategies are becoming more prevalent in matching DB liabilities and may contribute
to the investment strategy to ensure the product functions as intended.
The needs contained within the investment objectives to provide predictability in the future benefit payable to
members may result in investment strategies that are consistent with those observed in the more mature DB market.
Cashflow-driven investment strategies make greater use of hedging assets to reduce volatility, this in turn can deploy
dynamic discount rates which deliver greater funding stability.45
This approach necessitates a move away from
growth assets, which would have a consequential impact in the level of benefit offered as the price of offering greater
predictability.
Reducing growth assets and increasing the predictability of benefits would drive the benefit structure towards that
offered by an annuity which offers an insurer-backed guarantee of benefit levels. For a decumulation CDC product to
have a place in the market it must offer greater exposure to growth-seeking assets than used to back annuities, which
will inevitably lead to a degree of uncertainty in future benefit increases.
Introducing decumulation CDC to the pension market will make a
complicated decision more complex
For any potential new scheme member considering the market, or those guiding them, they will need to be able
to make valid comparisons between two schemes to understand how the schemes may suit their particular needs.
This necessitates engagement from the saver, yet a quarter of people have not accessed any information, advice or
guidance when planning for retirement. On accessing a pot, most people (71%) will have received information, advice
or guidance from a financial advisor, their pension provider or Pension Wise prior to accessing their savings,46
and it
will be important to consider where decumulation CDC is positioned within these markets.
The use of advice is associated with the purchase of more complex
retirement products.
When accessing pension pots for the first time to purchase an annuity or enter drawdown, nearly two thirds (64%) will
have received advice or guidance.47
There is a correlation between taking regulated advice and entering drawdown,
and receiving guidance through Pension Wise and purchasing an annuity [Figure 4.1]. While this does not imply
causation as there is a selection bias in the people who seek guidance or pay for regulated advice, it raises an issue
of how decumulation CDC should be positioned between the certainty of an annuity and the flexibility of an individual
drawdown product.
Figure 4.148
After advice people are more likely to enter drawdown
After guidance people are more likely to purchase annuities
Action on accessing pension pots for the first time after receiving advice or guidance when purchasing retirement
products, split by pot size (excludes tax free lump sum for annuity and drawdown products).
April 2020 – March 2021
0%
20%
40%
60%
80%
100%
Less
than
£10,000
£10,000
-
£29,000
£30,000
-
£49,000
£50,000
-
£99,000
£100,000
-
£249,000
£250,000
and
above
All
pot
sizes
Less
than
£10,000
£10,000
-
£29,000
£30,000
-
£49,000
£50,000
-
£99,000
£100,000
-
£249,000
£250,000
and
above
All
pot
sizes
Less
than
£10,000
£10,000
-
£29,000
£30,000
-
£49,000
£50,000
-
£99,000
£100,000
-
£249,000
£250,000
and
above
All
pot
sizes
With Regulated Advice With Pension Wise guidance With no advice or guidance
Proportion
of
pots
Full
withdrawal
Uncrystallised
funds pension
lump sum
Drawdown
Annuity
44
TPR (2023b)
45
Mercer (2023)
46
DWP (2022)
47
PPI analysis of data from Tables 10-13 of FCA (2022)
48
PPI analysis of data from Tables 10-13 of FCA (2022)
24
Prev Next

PPI - the role of CDC in decumulation

PPI - the role of CDC in decumulation

Recommended

Recommended

More Related Content

Similar to PPI - the role of CDC in decumulation

Similar to PPI - the role of CDC in decumulation (20)

More from Henry Tapper

More from Henry Tapper (20)

Recently uploaded

Recently uploaded (20)

PPI - the role of CDC in decumulation